International Research Journal of Engineering and Technology (IRJET)
Volume: 02 Issue: 07 | Oct-2015 www.irjet.net
CONTROLLING OF NONLINEAR SYSTEM BY USING FUZZY LOGIC
CONTROLLER
A.Vandana justice1, C.Kathyaini2
1 PG Scholar, Department of EEE, JNTU Anantapur, Andhra Pradesh, India
2 PG Scholar, Department of EEE, JNTU Anantapur, Andhra Pradesh, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - This paper presents the control of
nonlinear system by using fuzzy logic controller. In
order to synthesise a control system both the controller
selection and adjustment of parameters plays an
important role. PID controller is used in many
industries, It doesnt give acceptable response or
performance for system with uncertain time delays,
performance and nonlinearities. Hence the parameters
of PID controller are tuned to get satisfactory response.
Therefore tuning of PID controller is done by Fuzzy
logic. Fuzzy control establishes an intelligent control.
Fuzzy logic control has been widely used for
nonlinear higher order and time delay systems.
Because of their Knowledge based nonlinear structural
characteristics they are applied to nonlinear systems.
Fuzzy controller can perform online and offline
parameter operations. Fuzzy PID controller is designed
by fuzzy control principals and techniques. Fuzzy plus
PID provides good performance when compare to only
fuzzy. For fuzzy controller simple rule base is used,
whereas for Fuzzy PID different rule bases are used for
proportional, integral and derivative gains to make
faster response.
Key Words: Fuzzy logic controller, PID controller,
Nonlinear System, State Transition Algorithm, Genetic
Algorithm, Particle Swarm Optimization.
INTRODUCTION
PID controllers are the best known controllers
used in many industrial processes because of simple
structure and robust performance. PID controllers deal
with both time and frequency responses if they are tuned
properly. But for complicated systems the performance of
conventional PID controllers are tend to degrade. The
performance of the controllers changed drastically with
the introduction of fuzzy logic controller. For wide range
of industrial applications fuzzy logic controllers are being
used. This fuzzy logic controller shows better performance
when compared to PID controllers. The combination of
both fuzzy and PID produces good performance. Fuzzy can
2015, IRJET ISO 9001:2008 Certified Journal
e-ISSN: 2395 -0056
p-ISSN: 2395-0072
be applicable to any system. A fuzzy logic controller makes
control decisions by using If-Then rules.
After development of Fuzzy logic an important
application was developed in control systems known as
Fuzzy PID controllers. For solving nonlinear tuning
problems, the design of Fuzzy PID controller remains a
challenging area. For classic or modern control system
applications, instead of PID controller a Fuzzy PID
controller is used. They are used in converting the error
between the measured and reference variable in to
command. In practical design the information about input,
output transfer characteristics must be known. The main
purpose is to develop control system for all kind of
process with higher efficiency.
The introduction of Fuzzy block in to the structure of
linear PID controller resulted Fuzzy PID controller.
Equivalence between Fuzzy PID controller and a linear
control structure can be established by tuning method.
The both input variables and control action are given in
linguistic forms for Fuzzy PID controller. For conventional
PID controllers Fuzzy logic provides a certain level of
artificial intelligence. Fuzzy PID controllers provide self-
tuning ability and online adaptation to nonlinear and time
delay systems.
Fuzzy logic controller is classified in to two categories
named Mamdani and Takagi-Sugeno. For making decisions
mamdani type fuzzy logic controller uses fuzzy members.
Takagi-sugeno type fuzzy logic produces control actions
by linear functions of the input variables. Mamdani type of
fuzzy systems gives some tuning guidelines and
recommendations for increasing the quality of control
systems. This is used to solve the problem of tuning of
Fuzzy PID controller. For common structure of Fuzzy PID
controller the transfer function and equivalence relations
between controller parameters are obtained. A Fuzzy PID
controller is designed to increase the system response in
both transient and steady state when compared with
individual PID Fuzzy controllers. In Fuzzy PID controller
the parameters are tuned by using fuzzy logic controller in
order to get satisfactory response.
Fuzzy PID controller is designed with integral and
derivative control actions and set of 49 rules for
improving transient and steady state responses for both
first order and second order systems. For an industry
contains a typical PID controller design. Parameters of that
Page 648
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 02 Issue: 07 | Oct-2015 www.irjet.net p-ISSN: 2395-0072

controller are tuned manually to achieve desired plant

response. The manual tuning is replaced by fuzzy
controller. It improves the system response. Fuzzy control
system is an expert system applied to problem solving in
control.
By linguistic variables fuzzy rules provides natural
frame work for thinking and knowledge formation.
Selection of input and output scaling factors plays most
important role in design of fuzzy controller because of
their significant effect in controller stability and
performance. When compared to neural networks this
method avoid delays associated with network training.
There will be updating of rule base and membership
function.
1.PID CONTROLLER:
It is a control loop feedback mechanism used in
industrial control systems. PID controller is used to
calculate the error value as taking the difference between
the measured value and desired value. By adjusting the
control variables the controller minimizes the error
The three basic behaviour modes that are used by
the PID controller are P-proportional, I-integrate, D-
derivative. Proportional and integrative modes are used as
single modes whereas derivative mode is rarely used, but
combinations of PI and PD are commonly used. PID
controller is simplest possible on-off controller.
DESIGN OF PID CONTROLLER:
For estimation of nonlinear system model a PID
controller is designed for stability and other performance.
PID controller is widely used in industrial process.
The continuous form of PID controller is described as
follows:
u(t)=Kp[e(t)+1/Tit0 e(t) dt +Td d/dt e(t), (1)
Figure 1: The design of PID controller
yr (k) is the reference output
y (k) is the system output at the sampling point
PID controller is tuned to get fast response and to acquire
good stability.
2. FUZZY LOGIC:
Fuzzy logic is a multi-value logic which allows
some middle values define between conventional
evaluations like yes or no, high or low, true or false. It is an
expert system uses a collection of membership functions
and rules. Fuzzy system contains membership function
defined by the parameters. The degree of membership lies
between 0 and 1.Triangular, Trapezoidal, Gaussian are
membership functions.
The basic fuzzy model consists of fuzzification,
Interface, defuzzification and knowledge base. The input
to the system is crisp set. Fuzzification is nothing but
conversion of crisp set to fuzzy set. In fuzzification
membership function is applied and degree of
membership is determined. The fuzzy interference
process combines membership function with control rules
to get fuzzy output and these outputs are arranged in
tabular form called as look up table.
The process of producing quantifiable result in fuzzy
logic is called defuzzification. It converts fuzzy set in to
real numbers.

Where
e (t) is the desired error signal between desired and
actual outputs,
u (t) is control force
Kp is proportional gain
Ti is integral time constant
Td is derivative time constant

The approximations are followed:

0t e (t) dt T kj=0 e (j)

d/dt e (t) (e (k)-e (k-1))/T, (2)

The sampling time period is T.

Equation (2) can be rewrite as Figure 2: Block diagram of Fuzzy logic

u(k)= Kp{e(k)+T/Ti kj=0 e(j)+Td/T[e(k)-e(k-1)]}; (3)

2015, IRJET ISO 9001:2008 Certified Journal Page 649

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 02 Issue: 07 | Oct-2015 www.irjet.net p-ISSN: 2395-0072

3. FUZZY PID CONTROLLER: Ki : Integral gain

For tuning the parameters of PID controller some 4. STATE TRANSITION ALGORITHM:
conventional methods are carried out which have some
limitations. An intelligent method presented in this paper The process of tuning a system from one state to
is based on Fuzzy in order to tune the PID controller. If a other state is called state transition. It is described as state
system contains time delays, uncertain dynamics and transition matrix. State transition matrix has greater
nonlinearities, the conventional PID controller doesnt importance in modern control theory and also in
give acceptable performance. Hence the parameters of the communication theory. STA deals with state, state
PID controller are tuned in order to get satisfactory transition. A solution can be regarded as state and
response. Fuzzy logic is used to tune the PID controller updating solution can be treated as state transition. Four
automatically. Fuzzy system transforms a linguistic special operators are created to solve the optimization
control strategy in to an automatic control strategy based problems.
on expert knowledge. Figure 3 shows the block diagram of
Fuzzy PID controller. Fuzzy logic tool box in MATLAB is They are
used to implement Fuzzy PID controller. Rotational transformation ()
Translation transformation ()
The error e (t) and rate of change of error e are the Expansion transformation ()
inputs to the controller. While the outputs are controller Axesion transformation ()
gain Kp, Ki, Kd, its structure is to input, three output State transition algorithm is used to tune the PID
structures. The range of input and output membership controller parameters.
function is to be determined. The linguistic variables are
assigned as NB, NM, NS, Z, PB, PM, PS. The membership
functions used are Gaussian, Triangular, Trapezoidal and ALGORITHM:
Bell curves. Fuzzy rule base consists of If-Then rules,
which represents the knowledge level and human abilities 1. Repeat
which adjusts the system for minimum error and fast 2. If <min then
response. Fuzzy consists of two models named as 3. max
Mamdani and takagi-Sugeno model. Fuzzy control system 4. End if
is an expert system applied to problem solving in control. 5. Best Expansion (funcfcn, Best, SE, , )
6. Best Rotation (funfcn, Best, SE, , )
7. Best Axesion (funfcn, Best, SE, , )
8./fc
9. until the maximum iteration is met.
fc is constant coefficient.

5. GENETIC ALGORITHM:

Genetic algorithm is used to tune the parameters of the

PID controller. Genetic algorithm not only manipulates
one potential solution to a problem but a collection of
potential solutions. This is called as population. The
potential solution in the population is called chromosome.
These chromosomes are encoded by the parameters of the
solution. Each chromosome is compared by the other
chromosome in the population and rated according to
fitness. Population is also called as mating pool,
represented by binary or real coded values.
Creation of new chromosome from the existing
Figure 3: Fuzzy PID controller Block diagram
chromosome can be done by merging the existing ones or
modifying them. Reproduction can be done by Roulette
e: error signal
wheel method. The two genetic operators are cross over
e: change in error signal
and mutation. Cross over operator recombines the
Kd : Derivative gain
obtained information in to workspace. Mutation operator
Kp : proportional gain

2015, IRJET ISO 9001:2008 Certified Journal Page 650

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 02 Issue: 07 | Oct-2015 www.irjet.net p-ISSN: 2395-0072

generates new genes to ensure exploration. By this For Fuzzy PID controller:
premature convergence can be avoided.

ALGORITHM:

1. Initialize Population
2. Evaluate their fitness
3. Select the fittest member of the population
4. Reproduce using probability method
5. Implement crossover operation on the reproduced
chromosome
6. Execute mutation operation with low probability
7. Repeat step 2 until predefined convergence is met.

6. PARTICLE SWARM OPTIMIZATION:

Figure 4: PID controller design by using Fuzzy logic
Particle Swarm Optimization optimises the problem by controller for example 1
considering population of solution. The particles move in
the search space with some velocities. Each particle move For fuzzy PID
is influenced by the local best position. PSO is a pattern
search method which does not use the gradient of problem
that is being optimized. That means it does not require the
optimization problem. It is a simple algorithm. For a
number of iterations, a group of variables adjusted very
close to the member whose value is closest to the target at
any position.

ALGORITHM:

1. Initialise each particle

2. Calculate the fitness value of each particle
3. If the fitness value is better than the best fitness value
in the history, then set the current value as the new Pbest
4. Choose the particle with the best fitness value of all the
particles as the gbest.

SIMULATION RESULTS AND COMPARISION:
For controlling Nonlinear system a fuzzy logic controller
is utilised in order to give good stability and performance.
In order to implement Fuzzy PID controller fuzzy logic tool
box is used in MATLAB. In this paper the following to
insistences are studied.
Figure 5: Comparison graph between Fuzzy PID and
normal PID for example 1
Blue line represents fuzzy with PID which provides better
optimisation
Green line represents normal PID response that is not
better than Fuzzy PID

EXAMPLE 1: An unstable system is described by

x1 (k+1) = 1x1 (k)x2 (k), x1 (0)=1,
x2 k+1) = 2x12 (k) +u (k), x2 (0) =1,
y (k) = 3 x2(k) 4 x12(k),
The real parameters of the nonlinear system are assumed
to be = [1, 2, 3, 4] = [0.5, 0.3, 1.8, and 0.9].
The relative variables used in the Fuzzy PID are as follows
1[0,2], 2[0,2], 3[0,2], 4[0,2], N=8.

2015, IRJET ISO 9001:2008 Certified Journal Page 651

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 02 Issue: 07 | Oct-2015 www.irjet.net p-ISSN: 2395-0072

For State Transition Algorithm:

Figure6: PID controller optimization using Fuzzy logic

controller

Figure9: Kp, Ki, Kd graphs of STA

For Genetic Algorithm:

Figure 7: Kp, Ki, Kd simulation graphs for example

Figure 10: Kp graph of GA

Figure 8: FIS Editor of Fuzz PID

The model used here is Mamdani model. Figure 11: Ki graph of GA

2015, IRJET ISO 9001:2008 Certified Journal Page 652

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 02 Issue: 07 | Oct-2015 www.irjet.net p-ISSN: 2395-0072

EXAMPLE 2: Consider a first order with time delay

system whose transfer function is given as

G(s) =y(s)/u(s) = (k/Ts+1) e-ts,

K is the steady state gain
T is the time constant
t is the time delay with sampling time 0.01
The actual numerical values are K=10, T=5, t=9.

Figure 12: Kd graph of GA

For Particle Swarm Optimization:

Figure 14: PID controller design by using Fuzzy logic

controller for example 2

Figure 13: Kp, Ki, Kd graphs of PSO

Table 1: Best parameters of PID controller for

example 1

Algorithms Kp Ki Kd Figure 15: Comparison graph between Fuzzy PID and

Fuzzy PID 1.6409 1.118 2.09 normal PID for example2
STA 0.23 0.9 0
GA 0.15 0.34 0.1 Blue line represents fuzzy with PID which provides better
PSO 1 0.2814 0 optimisation

Green line represents normal PID response that is not

better than Fuzzy PID

2015, IRJET ISO 9001:2008 Certified Journal Page 653

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 02 Issue: 07 | Oct-2015 www.irjet.net p-ISSN: 2395-0072

The model used here is Mamdani model

Figure 16: PID controller optimization using fuzzy logic

controller

Figure 19: Membership function editor of Fuzzy PID

FOR STATE TRANSITION ALGORITHM:

Figure 17: Kp, Ki, Kd simulation graphs for example2

Figure 20: Kp, Ki, Kd graphs of STA

Figure 18: FIS Editor of Fuzzy PID

2015, IRJET ISO 9001:2008 Certified Journal Page 654

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 02 Issue: 07 | Oct-2015 www.irjet.net p-ISSN: 2395-0072

FOR GENETIC ALGORITHM: CONCLUSION:

In this paper a new optimization method named

Fuzzy logic controller is used to solve the problem of
controller design for nonlinear systems. As an optimizer,
Fuzzy logic controller is used to achieve accurate model,
and then it is adopted to obtain offline PID controller.
Fuzzy logic controller is used to tune the parameters of
PID controller in order to get good performance and also
to control the nonlinear system. Fuzzy with PID provides
better optimization when compared to other
optimizations. This optimization maximises the efficiency
of production. By comparison with STA, GA and PSO it is
found that Fuzzy PID is more stable. With regard to
convergence rate, it is also discovered that Fuzzy PID is
much faster than its competitors.

Figure 21: Kp, Ki, Kd graphs of GA

REFFERENCES:

1.Junhong Li, Rui Ding, Parameter estimation methods

for nonlinear systems, Applied Mathematics and
Computation, 219(9)(2013), 4278 4287
FOR PARTICLE SWARM OPTIMIZATION: 2. F.G. Shinskey, Process Control System: Application,
Design and Tun- ing, McGraw-Hill, 1996
.3..Holmblad L. P., Ostergaard J. J. (1982) Control of
Cement Kiln by Fuzzy Logic, Gupta M. M, Sarchez E,
Fuzzy Information and Decision Processes, North Holland,
pp. 389-399.
4. Meng Joo Er and Ya Lei Sun, Hybrid Fuzzy Proportional
Integral Plus Conventional Derivative Control of Linear
and Nonlinear Systems, IEEE Trans. On Industrial
Electronics, Vol.48, No.6 December 2001, pp. 1109-1117.
5. J. C. Wu and T. S. Liu, A Sliding-Mode Approach to Fuzzy
Control Design, IEEE Trans. On Control Systems
Technology, Vol.4, No.2, March 1996, pp. 141-151.
6. Wei-Der Chang, Nonlinear system identication and
control using a real-coded genetic algorithm, Applied
Mathematical Modelling, 31(2007) 541-550.
7. A Visioli of Diupartimento di Elettronica per
Figure 22: Kp, Kd, Ki graphs of PSO lAutomazione,Brescia University ,tuning of pid
controllers with fuzzy logic proceedings of ieee, Volume
TABLE 2: Best parameters of pid controller for 148, issue-1,Pages:1-8 , jan 20013
example 2 8. Bao-Gang, H.; Mann, G.K.I. & Gosine, R.G. New
methodology for analytical and optimal design of fuzzy
PID controllers, IEEE Trans. On Fuzzy Systems, Vol. 7,
ALGORITHM Kp Ki Kd
Issue 5, Oct. 1999, p. 521.
FUZZY PID 1.496 2.28 1.218
9. Hayward, G. and Davidson, V., 2003. Fuzzy Logic
STA 0.9 0.438 0.014
Applications, Analyst, Vol.128, 1304-1306.
GA 1.19 1.13 1.09 10. Alireza Al, Hamidreza Modares, System identication
PSO 1 0.24 0.05 and control using adaptive particle swarm optimization,
Applied Mathematical Modelling, 35(2011) 1210-1221.

2015, IRJET ISO 9001:2008 Certified Journal Page 655

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 02 Issue: 07 | Oct-2015 www.irjet.net p-ISSN: 2395-0072

BIOGRAPHIES

A.Vandana justice currently pursuing

M.Tech Control Systems at JNTUA College Of
Engineering and Technology Anantapur,
Andhra Pradesh. Her area of interest is
Control Systems.

C.Kathyaini currently pursuing M.Tech

Reliability Engineering at JNTUA College Of
Engineering and Technology Anantapur, Andhra
Pradesh.
Her area of interest is Reliability Engineering.

2015, IRJET ISO 9001:2008 Certified Journal Page 656